PRTG Network Monitor 8 Review

20 Oct

The other day I was checking my emails and amongst the mundane rubbish of people trying to flog me Viagra or robots spamming my blog, I spotted an interesting mail from a person at the PR company for Paessler, the company that make PRTG Network Monitor. Basically, they had read the post I had written on PRTG 7 mapping a while back and wondered if I would like to take a look at the latest version of their software. Having already seen a few interesting bits and pieces on the Paessler website about version 8, I promptly agreed.

I’ve been running version 7 for a little while now and it’s pretty awesome. (I actually have a seperate PC and monitor on my desk with chrome running in fullscreen and 3 tabs with various maps and alerts from PTRG on there. I then have a script to swap between the tabs every 10 seconds. I also control this PC with InputDirector.) I have roughly 300 sensors, monitoring everything from bandwidth on the main firewall, to Blackberry services on the BESx box, desktop PC statuses, content checking on our public websites, monitoring of SQL and MySQL databases on remote servers using probes, all in all, quite a lot. I normally also add a few more sensors each week on top of this lot. I recently added WMI network monitoring for each desktop in the office, I then added ‘Channel’ settings for the network card and set traffic in and out limits. This way, if there is heavy traffic on the LAN/WAN, I can visually see where it’s coming from, or rather who is responsible (The Traffic Shaper on the PFSense box makes sure my bandwidth is all evenly allocated and all critical serices are protected, but I like to keep an eye on things anyway.)

As I already had V7 running (on Server 2K8), I didn’t opt for a fresh install, but instead I upgraded the existing box. It was on a Hyper-V VM so I took a backup of the VHD before I started, just in case. The installation was a breeze and kept all my existing setting intact. All the existing remote probes worked with V8 and so I haven’t really had to change too much. My maps, and everything else, just look a bit more updated but worked perfectly well still.

Before…

And after…

For those doing a fresh install, there might seem like and overwhelming amount of sensors and setup involved, but it’s really very simple. Basically, you want a server running the Core PRTG Network Monitor, this can monitor anything it can see on its local network, or even extended over VPN’s. You can easily add sensors by entering the IP address, or running the ‘Auto-Discovery’ feature. You can then add in remote probes (a small service you can install on remote, hosted servers for example). All you need to do then is NAT TCP port 23560 over to your Core server on your firewall and then you can monitor whatever the probe(s) can see too. (I also created an external DNS entry and natted TCP port 80 so that I can login and check the status of everything easily remotely.) You can then access PRTG Network Monitor either via the webGUI or by installing the Windows GUI (from the webGUI). I personally prefer the WebGUI using Chrome, but each to their own.

The one big change that is most obvious is the licencing, it’s now much much simpler. The free version allows up to 20 sensors (or 10 if you don’t add a free link to Paessler on your website) and then the commercial version (which allows between 100 and an unlimited number of sensors!) is split into the following price brackets:

You can also trial up to 5000 sensors for 30 days, click here for more information.

I don’t currently own an iPhone (in fact I’m between contracts right now so if anyone from Apple is reading this and wants some free promotions, an iPhone 4 would do nicely) but the iPhone app from Paessler does look pretty smart and is available for download for just $11.99 from here. Paessler have also created a “Mini-HTML” interface version of the website, to run on Android and other smart phones (Shame I don’t have a HTC Desire HD either…).

There are a lot of improvements over version 7, a lot of the aesthetics have changed, this might be quite sad to admit, but I really prefer the new status box icons! It’s also now even easier to use in my opinion and gives for much greater control of your network. There’s a whole heap of Linux monitoring stuff (which I won’t be looking at just yet), some NTP and LDAP monitoring (which I have added and they work very well), a WMI uptime sensor which is pretty handy, a whole load of VMware monitoring, Cluster Statuses, Packet Sniffing (which I plan on using later this week) and a stack more.

It also worked well in all browsers, but if you’re having trouble with Internet Explorer, just check that you’re not in Compatability mode.

The use of AJAX is very impressive, when creating sensors for example, you can start typing the name and it will populate the list of matching sensors immediately, making it even easier and quicker to get setup and running. There’s also a lot of mouseover type stuff which makes navigation a lot easier, well, it saves on having to click buttons, thus, saving me time and energy.

So that’s the basics all covered, now let’s look at the cool stuff, and there’s a lot of it.

Firstly, check out the awesome icons available to you when creating maps, these kinda put my previous VISIO maps to shame and so I’m going to have to re-create all my network maps now! It just so happens that I’m retiring a few of our externally hosted servers in the next few months, so when the new ones are up and running I’ll be sure to create some new maps with added awesomeness (There will be a blog update, watch this space!)

Next, maps. Sounds simple, and it kinda is, but it’s also pretty sweet. In the settings for each group you can specify the Location (Postcode or Address) and it will stick in the relevant Google Map on the sensor/group page.

Once you have entered the GeoMap Data you can then select a Geo Map Object to add to you maps, pretty sweet. So now you have all your core servers and services grouped by location as well as category and so when a problem hits you, you can visually see what’s going on and where, very quickly.

Depending on which section you are viewing, the Google map will show the relevant geographical location(s) and their status.

The last bit off coolness, which I think you could possibly do with version 7, but it’s nice all the same, is adding all sorts of extras into your maps, like CCTV and weather etc. there’s a nice guide here on how to add your own HTML code in to your maps. You can also grab pretty much any widget you like from here and stick it in your maps. I haven’t really come across anything else which is of any use, but feel free to let me know if you do!

I forgot to point out in my previous post that there are some quite cool add-ins available from the Google code site, here.

Just a few of them that I think are pretty handy are:

  • CheckDHCP – Returns the response time of a DHCP server.
  • SSLCertExpiration – Returns the number of days before your SSL certificate expires.
  • UserLoggedin – Returns the number of users logged in on a computer.
  • WinOSVersion – Gets the Windows OS version of the specified computer.

For a guide on using custom sensors, see the video here.

Anyway, all in all, this software is great, and if you’re not running it, frankly there is something wrong with you. I can’t really say a bad word about it to be quite honest. Version 7 worked well enough and with all the latest additions, version 8 is even better. I’m not even sure what I could possibly need extra from version 9, but I’m sure they’ll think of something.

I’ve really only scratched the surface with what this software can do. I could spend all day writing about it, but I’m sure you get the idea and I guess you’re now wondering how you ever worked without it.

Actually, there was something I was after, I wanted to check the toner levels on a Konica-Minolta BizHub C451 and report on this but despite quite a few hours trying various custom SNMP OID Values, I really didn’t get anywhere. If anyone has any ideas on this, please let me know. I also need to investigate how I might check the status of the RAID arrays on my various servers, I’m currently using the HP event notifier, but I’m hoping to link this into PRTG down the line so that all my monitoring is in one place.

Update: I forgot to mention about adding LDAP sensors. For this you need to use a domain account and provide the DN and password for this account. The easiest way I found to discover the DN was to run this:

DSQUERY.exe * -limit 0 -filter “(&(objectCategory=Person)(objectClass=User)(!userAccountControl:1.2.840.113556.1.4.803:=2))” >”c:\Myusers.txt”

from the command line on a domain controller. Then copy the text inside the quotes for the user that you wish to use to check LDAP, along with the matching password of course.